On 8/29/06, Bob Axtell wrote: > Peter Todd wrote: > > On Sun, Aug 27, 2006 at 04:35:58PM -0400, David VanHorn wrote: > > > >>> There's an idea for you Peter, take a traditional clock and replace the > >>> motor with a stepper. Have it work normally keeping time but > >>> periodically "go nuts" backwards, fast forward etc, then correct itself > >>> and behave for a random period.That might sell.. > >>> :) D > >>> > >> A PIR detector could be used to detect when you were really in front of it > >> "watching" so as to prevent the behaviour during those times, but allow it > >> to go nuts when nobody's all that close to it. > >> > > > > That's a pretty cool idea! Actually a friend of mine made an art peice > > using a similar concept, she built various types of playground > > equipment, see-saw, swing and one of those merry-go-round things, and > > all made them activate and move while you were busy reading her artists > > statement. They stopped the second you moved away to look at the work. > > > > In practice of course, standing near the work, but looking away, still > > kept the PIR sensor going... So... > > > > How would you go about detecting the someone's gaze? > > > > Personally I'd rig up a digital camera, with "eye-detecting" software, > > but surely there are better ways... > > > > > The problem is that most very sensitive PIR detectors/algorithms work > when the warm object is MOVING. > You might need to design your own (all of mine were quite sensitive > (40m), What if you have a thermal noise like a radiator in front of it ? but required the target to move > every few seconds or so. Put a mechanical chopper in front of the PIR detector and the subject can stay... easy, isn't it ? greetings, Vasile -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist